“The Police and the Populace”: Canadian Media’s Visual Framing of the 2010

Aziz Douai University of Institute of Technology

AbSTrACT The clash between the slick marketing slogans of the police and the democratic protections of political dissent was on full display throughout citizen protests during the 2010 G20 meetings in Toronto, . In addition to the summit’s excessive costs and organiza - tional lapses, the eruption of violence and questionable police tactics dominated media cov - erage of the summit. This research investigates the media’s visual framing of the policing of the G20 Toronto summit through an analysis of 852 news images published in several print and online media outlets in Canada. The article examines how the “visual tone” of the images, news ideology, and the news medium affect the visual framing of the anti–corporate global - ization movement in communications research.

KEywOrDS Visual communication; Frame analysis; Ideology; ; New media rÉSUMÉ Le conflit entre les slogans mercatiques sophistiqués de la police et les garanties démocratiques de la dissidence politique a été dévoilé lors des affrontements violents entre la police et les protestants lors des réunions du G20 en 2010 à Toronto au Canada. En plus des coûts exorbitants et les faiblesses organisationnels, l’éruption de la violence accompagnée des tactiques policières qui ont été mises en question ont dominé la couverture médiatique durant le sommet. Cette étude examine le cadrage visuel du maintien de l’ordre et la sécurité par les médias au sommet du G20 à travers une analyse de 852 images d’actualité publiées dans plusieurs journaux ainsi que des médias électroniques au Canada. L’article étudie le « ton visuel » des images, idéologie des medias, et le moyen d’information médiatique affectant le cadrage visuel du mouvement d’opposition au mondialisme.

MOTS CLÉS Communication visuelle; Analyse du cadrage; Idéologie; Mondialisation; Nouveaux médias

Introduction ommunications media have long constituted a symbolic battleground on which Ccompeting social constituents struggle to “frame” and present coherent narratives (Entman, 1993; Gitlin, 1980), control the pictures in people’s minds (Lippmann, 1922), vie for greater “visibility” (Thompson, 2000), and lobby “public opinion” to their side of history. In this contentious battle over power in society, corporate images, pub - lic pronouncements and outreach, strategic communication, and public relations

Aziz Douai is Assistant Professor of communication, Faculty of Social Science and Humanities, at the University of Ontario Institute of Technology, 2000 Simcoe Street N., Oshawa, ON L1H 7K4. Email: [email protected] .

Canadian Journal of Communication Vol 39 (20 14) 175 -192 ©20 14 Canadian Journal of Communication Corporation 176 Canadian Journal of Communication, Vol 39 (2) have led the way. while strategic communication refers to “the strategic application of communication and how an organization functions as a social actor to advance its mission,” (Hallahan, Holtzhausen, van ruler, Verčič, & Sriramesh, 2007, p. 7) corpo - rations have long cultivated their brands and passed themselves to the public as so - cially responsible corporate citizens. Non-governmental organizations employ strategic communication tools to attract public support for their myriad environmen - tal, social, or political causes. Governments around the globe are positioning them - selves as corporate enterprises that “sell” policies to the larger public (Grunig, Grunig, & Dozier, 2002). Like other state organizations, police services are no different in try - ing to manage their public image carefully and glean public support from which they draw their legitimacy in liberal democracies (Mawby, 2002). For instance, the ’s slogan, “To Serve and Protect,” has the unmistakable ring of a slick marketing slogan. For the public at least, the slogan encapsulates the mission of the police, which casts it in peaceful-sounding terms, “To Serve.” On the other side of the spectrum, the anti–corporate globalization movement has both appropriated and subverted these forms of “self-mass communication” to fight for an equitable world (Juris, 2005; Kahn & Kellner, 2004). The clash between the police department’s slick marketing slogans and the dem - ocratic protections of political dissent was on full display throughout the violent con - frontations and the policing of citizen protests during the 2010 G20 meetings in Toronto, Canada. The massive security operation of the G20 summit became the largest in Canadian history, involving local and provincial police, the rCMP, and the military (Monaghan & walby, 2012). In addition to the unprecedented disruption of normal life in the city, the summit’s excessive costs and organizational lapses, the erup - tion of violence and questionable police tactics eclipsed media coverage of other sum - mit issues. News images of burning police cruisers, threatening depictions of so-called protesters, and coverage of Toronto police rounding up potential suspects and incarcerating them in makeshift prisons overwhelmingly dominated media re - ports (Monaghan & walby, 2012). This article investigates the media’s visual framing of the policing of the 2010 G20 Toronto Summit through an analysis of 852 news im - ages published in mainstream print and online media outlets in Canada. Employing a visual framing analysis, the article analyzes how the visual “tone” of images, news organizations’ ideological leanings, and the news medium influence the visual frames of the anti–corporate globalization movement in the west. Covering the anti–corporate globalization movement Police and anti–corporate globalization protesters have waged their communications war since the late 1990s, attracting researchers’ attention to the raging battle of images. From the “battle of Seattle” to “the battle of Genoa,” existing research related to the anti-globalization movement during the (wTO), G8, or G20 meetings has almost exclusively focused on the communicative styles and strategies of the dissenting populace (e.g., Juris, 2005; Kahn & Kellner, 2004; wall, 2003), or the media and journalists’ reporting strategies and framing of dissenting protests (e.g., Hall & bettig, 2003; Jha, 2008), ignoring the state-apparatus and police forces’ communication tactics. For instance, wall (2003) used the participant-observation Douai “The Police and the Populace” 177

method to analyze and compare the communication tactics non-governmental organizations and the decentralized “street movement” groups employed during their 1999 protests against the wTO known as the “battle of Seattle.” wall (2003) concluded that the “street movement” groups employed a “radical frame,” unlike the institutional or “reformist frame” (p. 44) used by NGOs seeking to influence mainstream media. Juris (2005) critiqued the media-circulated images of wanton violence and destruction emanating from the anti–corporate globalization protests during the G8 summit in the Italian city of Genoa in 2001. He further argued that Genoa police successfully exploited images of protesters’ own “performative violence” (p. 414) to portray militants as a threat to the social and moral order. Despite different organizational covers, anti–corporate globalization protesters are united by their conviction that rich countries’ policies “enable the super rich and pow - erful to exploit workers, ignore human rights, squeeze third world countries, and de - stroy the environment—all in pursuit of profits” (Hall & bettig, 2003, p. 3). However, these issues never constitute the central “frames” in the presentation and coverage of the anti–corporate globalization movement by mainstream media (MSM). All too often, MSM’s portrayals marginalize demands of activists by promoting narratives of “law and order” (Juris, 2008). Media ownership structure and framing strategies have been blamed for the jaundiced media coverage of dissent. In The Whole World Is Watching , Gitlin (1980) examined how the anti– war movement of Students for a Democratic Society (SDS) was marginalized and portrayed as a threat to the social and moral order of the . The mainstream media “framed” SDS members as terrorists and anarchists bent on destruction, leading to the fragmentation and ul - timate failure of SDS. Dominant and biased media frames largely result from concen - trated media ownership in the hands of few corporations, which use their media power to serve and perpetuate the interests of the dominant structure. As Herman and Chomsky (1988) argue, the media are organically embedded in the dominant power structure, essentially both belonging to the same elite, and as such are disinterested in presenting counter-hegemonic stories that may pose a threat to U.S. “national inter - ests.” In the “propaganda model,” Herman and Chomsky critique the mainstream media’s overwhelming reliance on “official” sources—instead of oppositional voices— and on advertising whose ideological biases constitute “filters” that lead to the deni - gration of political dissent. In contrast, the Internet has become a boon for dissenting social movements and activists seeking to circumvent MSM’s skewed gaze in both distributing and receiving alternative news. while the role of the Internet and new media in energizing social movements cannot be underestimated (Kidd, 2003), most studies focus on social movements’ use of the Internet as a “mobilization” and “counter-framing” tool. The rise of what Castells (2007) describes as “self-mass communication” opens up new spaces for more “horizontal communication” (p. 239) from the many-to-many. In the “network society,” anti–corporate globalization activists and other social actors seek to harness these “self-mass communication” digital media tools to create a new “counter-power,” that is, new tools to “challenge and eventually change the power re - lations institutionalized in society” (p. 248). The success of the Indymedia and 178 Canadian Journal of Communication, Vol 39 (2)

Independent Media Center projects in offering alternative news that challenges the hegemonic narratives of corporate media demonstrates the vital role of Internet ac - tivism (Kidd, 2003; Pickard, 2006). yet, despite this promise, the websites of main - stream media still retain an audience reach advantage. Their capacity to “mass communicate” on these online platforms remains unrivalled due to their established institutional and organizational advantages over alternative news outlets (Hindman, 2008). The need to examine mainstream media’s online platforms is as great as the task of scrutinizing what Hindman calls “the myth of digital democracy” (p. 81). That is why this study compares both print and online coverage by mainstream media of the anti–corporate globalization movement. Framing and the “visual turn” Media researchers have long recognized the media’s influential role in shaping the so - cial world, processes described as “priming,” “agenda setting,” and “framing” (Scheufele & Tewksbury, 2007). As a theory of how news stories are “constructed,” framing embodies a “visual trope” in communication scholarship (Frosh, 2011) that implies the existence of “prisms,” “windows,” and “angles” through which audiences make sense of news events. Frame construction involves “selection, emphasis, and presentation” (Gitlin, 1980, p. 7) of certain aspects of the news to increase their salience as “interpretative” frameworks. A similar emphasis underlies Entman’s (1993) defini - tion of framing as the selection of some aspects of reality “to make them more salient in a communicating text in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation” (p. 52). The fact that “selection and salience” symbolically promote certain narratives has been captured in reese’s (2001) definition of frames as “organizing principles that are so - cially shared and persistent over time, that work symbolically to meaningfully structure the social world” (p. 11). Since words, phrases, and images are indispensable tools used to construct these “organizing principles,” or central narratives, media researchers have focused not only on the frames’ effects on audiences, but also on how these frames are constructed in the first place. The preoccupation with the construction and production of frames in media discourse, however, seems to have largely been limited to the traditional “textual” aspects of news stories, notwithstanding a deeper appreci - ation of the impact of visuals on the larger news frames. The centrality of news images to the news framing process has recently been ac - knowledged in what has been described as a “visual turn” (Jay, 2002) in communication research. According to Griffin (2004), “more than they describe, photographs tend to symbolize generalities, providing transcending frames of cultural mythology or social narratives in which the viewer/reader is led to process and interpret other information on the page or screen” (p. 384). Scholarship on visual frames has begun to narrow this gap in framing research (e.g., Messaris & Abraham, 2003). In a comparative analysis of and the Guardian newspapers’ photographs of the 2003 Iraq war, Fahmy and Kim (2008) concluded that the pictorial coverage was “unprecedented” (p. 455) in its emphasis on the human cost of the war, particularly among Iraqi civilians. News photographs assist audiences in cultivating not only awareness but also recollec - tion of momentous events, and thus shape a nation’s collective memory (Sontag, 2004). Douai “The Police and the Populace” 179

In an investigation of visual frames of the first, second, and third anniversaries of the Iraqi invasion as a “past (resurrected) event,” Schwalbe (2006) found that these visual frames emphasize “the human face of war” rather than the war machine itself. Despite the productive and systematic nature of framing analysis, media re - searchers have had to confront a persistent question: How can media frames be ana - lyzed and detected in a way that yields “valid” and “reliable” (i.e., replicable) results? Thus, content analysis of visual data is quite common in visual framing studies that “inductively” or “deductively” identify the visual frames in photographs (e.g., Fahmy, 2004, 2007; Messaris & Abraham, 2001; Schwalbe, Silcock, & Keith, 2008). For instance, Schwalbe (2006) preferred to inductively glean visual frames from the actual photo - graphs of the Iraqi war, rather than analyze the visual frames depicted in the data de - ductively. Similarly, Parry’s (2010) analysis of british press photography used a similar “inductive” approach and then had independent coders code the data for content analysis. In this study, an inductive approach is used to analyze the visual frames found in news photographs of the Toronto G20 protests. The benefits of this inductive ap - proach include a close and data-driven analysis that permits “visual frames” to emerge directly from the data. Research objectives and questions The production and dissemination of news images of violence in anti–corporate glob - alization protests constitutes an important subject of study because of their potential to de-legitimize dissent in liberal democracies. Several interrelated objectives warrant this inquiry into how the media’s visual framing strategies constructed the anti–cor - porate globalization protests during the Toronto G20 summit. The overarching goal of this study was to determine how news images of protesters’ “violence” during the Toronto G20 summit fit within the overall police communication strategy that deni - grates and demonizes activists and protesters. In terms of communication scholarship, this research contributes to framing literature regarding the potential intersections be - tween the media’s visual and textual frames. Another objective relates to tracking the evolving impact of mainstream new media on the visual framing of social movements. The last goal is articulated through comparing the visual frames circulated in main - stream print and online media outlets. In approaching these objectives, the first research question of the study (rQ1) ad - dresses print media’s visual “tone” regarding anti–corporate globalization protestors, the police, and world leaders. Specifically, how do “primary” news images compare with “secondary” news images in the “tone” of their visual coverage of the G20? The visual “tone” of news has been found to play an important part in the framing of social protest and politics (e.g., De Vreese, Claes, banducci, Semetko, Hall, & boomgaaden, 2006; Fahmy, 2007; Dimitrova, Kaid, williams, & Trammell, 2005; McLeod & Detenber, 1999). News photographs’ visual “tone” (i.e., “neutral,” “positive,” or “negative” frames) constitutes an important part of the overall news frames that can be used to either le - gitimize or de-legitimize political expression and dissent. A “negative” visual tone can imply that protesters represent a threat to society, whereas a “positive” tone regarding the police can overemphasize their role as “guardians” of social peace, and thus may even legitimize their excessive use of violence and force. based on accumulated litera - 180 Canadian Journal of Communication, Vol 39 (2)

ture on how mainstream media deal with political dissent and the G20 protest move - ment (e.g., Gitlin, 1980; Hall & bettig, 2003), it is expected that the data will confirm a negative framing of protesters, and a positive framing of the police and world leaders. Since this is an inquiry into news images, comparing the visual frames emanating from “primary” and “secondary” news will be conducted to offer a more nuanced under - standing of the “visual tone” of news coverage. The ideological leaning and ownership of print media must be interrogated to fully grasp the impact of visual framing. In this regard, the second research question of the study (rQ2) investigates how the “ideology” of mainstream media affects the visual framing of social protest. Specifically, how does the ideological leaning of news organizations influence the visual framing of anti–corporate globalization protesters, the police, and world leaders? below, I outline three overarching visual frames that dominate media coverage of citizen activists’ encounters with police during these G20 meetings: a “lawlessness and violence” frame, a “law and order” frame, and a “civil liberties” frame. To address rQ2, I compare how different newspapers with divergent ideological leanings employ each of the above visual frames. based on the proliferation and growing influence of new media, I further compare the visual framing of the G20 protests by studying how images from print and elec - tronic news sources portrayed world leaders, the protesters, and the police. Hence, the third research question (rQ3) compares print and online mainstream media’s visual framing of anti–corporate globalization protesters, the police, and world leaders. Focusing on CbC.ca and CTV.ca, the study examines these websites as examples of mainstream online media in Canada. New media’s ascendance requires a framing analysis of published news images to shed light on how Canada’s mainstream media deploy visual framing strategies and various media platforms to legitimize or de-legit - imize political dissent and anti–corporate global activism. Method and data analysis procedures This research examines mainstream media’s print and online visual framing of the G20 Toronto summits. Canada hosted the 2010 G8 Summit in Huntsville, Ontario, on June 25 and June 26, while the 2010 G20 Summit was held in Toronto on June 26 and June 27 (TPS, 2011). To better capture the media’s visual “tone” during this period, the study’s sample focuses on news images published in Canadian media outlets between June 24 and June 28, 2010, a day before and a day after the official summit dates. This was a period of intense media attention given the chaos that reigned over Toronto and the ensuing violence as well as the questionable police interventions and reaction to anti–corporate globalization protests. To assess how Canadian media visually framed the G20 events, newspaper images were retrieved from microfiche newspaper records. In addition, news images of the G20 published on CbC.ca and CTV.ca were retrieved from the websites between May and June 2011. A second search of these websites was conducted in June 2012 to ensure that any missed images were included in the data. The study selected eight Canadian newspapers and two mainstream news media websites based on their position and status in the Canadian news ecosystem. The analy - sis focused on the print editions of selected newspapers and the websites of two main - stream Canadian broadcasters, CbC.ca and CTV.ca. Newspaper issues were ordered Douai “The Police and the Populace” 181 from Library and Archives Canada. Geographic location, national and regional reach, corporate ownership, and political influence were the primary criteria used to select the newspaper sample (see Table 1). In terms of location, the sample included major publications from four Canadian regions (Atlantic, Central, the Prairies, and the Pacific). The sample represented diverse ownership structures and affiliations, from Torstar, Postmedia, and Quebecor to CTVglobemedia. while the and are national dailies, the rest of the sample includes a mix of local and regional newspapers. The sample exhibited diversity at the level of circulation, with the reaching a weekly circulation average of more than 2 million, whereas St. John’s Telegram represented the least circulated newspaper in the sample. Le Journal de Montréal and La Presse , two influential French-language newspapers based in Québec, were also included. Finally, publication frequency was taken into considera - tion: about half of the newspapers in the sample are published from Monday to Saturday. This was an important consideration given that the G8 and G20 events took place over the weekend of June 24, 2010. The second research question focuses on how the ideological affiliation of newspapers influences visual frames. Although it is chal - lenging to categorize the ideological affiliation of news outlets, scholars can usually look to the editorial political leanings and ownership structure to get an approximate ideological categorization. Applying these tools to the Canadian mainstream news media, the Toronto Star has been considered as “left-of-centre” because of its “social-

Table 1: Selected newspapers

Weekly Canadian Publication Newspapers Language Province Average Owner Region Frequency Circulation

B.C. & Postmedia The Sun English B.C. 1,072,029 M-Sa Yukon Network Inc.

Winnipeg Free Press English Prairies Manitoba 891,133 M-Su F.P. Cdn. Np Ltd.

The Telegram, Newfound- Transcontinental English Atlantic 138,876 M-Su St. John's land Inc.

Power Corp. La Presse, Montréal French Québec Québec 1,276,623 M-Su of Canada

Le Journal Quebecor/ French Québec Québec 1,469,899 M-Su de Montréal Sun Media

Toronto Star English Ontario Ontario 2,044,024 M-Su Torstar

CTV The Globe and Mail* English Ontario Ontario 1,906,686 M-Sa Globemedia Inc.

Postmedia National Post* English Ontario Ontario 949,498 M-Sa Network Inc.

* While they are Ontario-based, the Globe and Mail and the National Post have national circulation and are thus classified as national dailies. 182 Canadian Journal of Communication, Vol 39 (2) liberal editorial stance” (Greenberg, 2000), the Globe and Mail represents the main - stream political “center” (Jiwani, 2005), while the National Post falls within the “right wing” Canadian political spectrum because of its unabashedly conservative political stance (Greenberg, 2000). Although the foregoing does not constitute a full ideological analysis, this article will explore the relationship between the ideological bent of a newspaper and its visual framing of social protest. Image data The image was the unit of analysis. A total of N=852 G20-related images were catego - rized as either “primary” or “secondary” images. “Primary” images refer to large-size photos prominently placed on the newspapers’ sheets and/or accompanying a main news story, while all smaller-size images were coded as “secondary.” In addition, all images at the top of the news item in a news website, and plainly visible without scroll - ing down, which is roughly the equivalent of “above the fold” newspaper stories, were coded as “primary” images. Using microfiche archives of the selected newspapers, a total of 386 were retrieved from the issues of eight newspapers. Of these, 162 images were coded as primary and 224 were coded as secondary (see Table 2).

Table 2: Print media image sample

Number of Number of Grand Newspapers Primary Images Secondary Images Total

Vancouver Sun 17 12 29 Winnipeg Free Press 9 11 20 The Telegram, St. John’s 4 4 8 Le Journal de Montréal 13 11 24 Toronto Star 56 88 144 La Presse, Montréal 7 6 13 Globe and Mail 29 22 51 National Post 27 69 96 Total 162 224 386

Similar to the print media sample, the online media sample represented a mix of ideological and professional affiliations in Canada’s mediascapes. Two news websites belonging to Canada’s most influential broadcasters were included in the study: the Canadian broadcasting Corporation’s (CbC) news ( www.cbc.ca ) and CTV News (www.ctv.ca ). These websites were included because they are well-established news outlets in the Canadian news ecosystem. Using the keywords “G8/G20 Toronto sum - mit,” “G8,” and “G20,” these websites were searched and yielded many images pub - lished between June 24 and 28, 2010. In addition, hyperlinks to other stories published on the websites were traced to ensure that all related images were included in the data. A total of 45 images were retained from the two websites, including 28 “primary” im - ages and 17 “secondary” images (see Table 3). Douai “The Police and the Populace” 183

Table 3: Online media image sample

Number of Primary Number of Secondary Grand Websites Images Images Total CBC.ca 18 8 26 CTV.ca 10 9 19 Total 28 17 45

Coding categories A visual analysis of the data was conducted following a simplified coding sheet at two stages. The first stage focused on the visual “tone” of media coverage, particularly how anti–corporate globalization protesters, the police, and world leaders were portrayed. Lack of access to hard copies of the newspapers selected for analysis prevented the re - searcher from a better measurement of the images’ size or employment of additional coders to calculate “intercoder” reliability. To address this shortcoming, the researcher showed a sample of G20 news photographs to media studies faculty and students to gauge how they would describe the visual impact and message of these photographs. based on the feedback, image portrayals were coded as “positive,” “negative,” or “neu - tral” in tone. “Positive” images refer to those that “glamorize” their subjects, be they police, protesters, or world leaders. If police and other members of law enforcement were portrayed glamorously while preserving “law and order” and carrying out their duties, their image would be coded as “positive” (e.g., Figure 1). A “positive” image of anti–corporate globalization protesters would be one that shows protesters as march - ing in a peaceful or orderly manner. A “negative” image portrays its subjects as violent or threatening, or depicts desolate background scenes and disorderly gatherings. If an image shows a chaotic scene, where protesters are breaking store windows, for instance, it would be coded as “negatively” framing the protesters (e.g., Figure 2). Images that fail to clearly criticize or glamorize their subjects were coded as “neutral.” Thus, news photographs that neither criticize nor praise the police, activists, or G20 world leaders were coded as “neutral” in tone (e.g., Figure 3).

Figure 1: A squad of mounted police patrol during the 2010 G20 Summit in Toronto

Source: CBC.ca 184 Canadian Journal of Communication, Vol 39 (2)

Figure 2: A protester vandalizes a storefront window during the 2010 G20 Summit in Toronto

Source: CBC.ca

Figure 3: World leaders in the 2010 G20 Summit meeting

Source: CBC.ca

In the second stage of the analysis, I consulted previous research on the anti–corpo - rate globalization movement, protesters, and police violence, as well as mainstream media’s coverage of previous G20 meetings. In keeping with previous research, my review concluded that three overarching frames prevail in media coverage of citizen activists’ encounters with police during these G20 meetings. These frames are a “lawlessness and violence” frame, a “law and order” frame, and a “civil liberties” frame. Images with a “lawlessness and violence” frame highlight the violence and chaos that ensued after the police and protesters clashed. For instance, images of a police cruiser on fire, overt use of violent force by the police , or of protesters breaking a store window would be coded as highlighting the “lawlessness and violence” frame (e.g., Figure 2). A “law and order ” frame, however, would emphasize the need for security and the “constructive” role police forces play in ensuring citizens’ safety. For instance, images that highlight the “security zone” erected by the police, or show the police in full riot gear, would be considered part of this frame (e.g., Figure 5). The “civil liberties” frame would include all news photo - graphs that emphasize citizens’ democratic rights to dissent and free speech. For instance, this frame would be depicted in an image of citizens using creative ways to express their disapproval of the G20 economic and social policies, or one that portrays Aboriginals Douai “The Police and the Populace” 185

protesting against corporate-driven globalization, or an image that depicts demonstrators marching for maternal health and women’s rights issues (e.g., Figure 4). In some cases, images impart multiple frames, a “law and order” and “civil liberties” frame (e.g., Figure 5), and those cases were coded as featuring multiple frames.

Figure 4: Activists march for women’s health during the 2010 G20 Summit in Toronto

Source: CBC.ca

Figure 5: A protester and police during the 2010 G20 Summit in Toronto

Source: CBC.ca Results Visual tone The first research question analyzes how newspaper images of the G20 depicted anti– corporate globalization protesters, the massive police forces deployed for the meetings, and world leaders attending the G20 meetings. Images that do not depict protesters, the police, or world leaders were labelled as “other.” Such images included but were not limited to deserted neighbourhoods, security fences, world leaders’ spouses, bro - ken storefronts, or desolate buildings. Moreover, the study distinguished “primary” images that dominate the page from “secondary” images, those of lesser prominence. Image depictions of protesters, the police, world leaders, or “other” were coded as “positive,” “negative,” or “neutral” (see Table 4). 186 Canadian Journal of Communication, Vol 39 (2)

Table 4: Visual coverage of G8 and G20 summits in Canadian print media

Primary Images Secondary Images

Negative Positive Neutral Negative Positive Neutral Total

% N % N % N % N % N % N % N

Protesters 26.87 36 11.19% 15 11.19% 10 33.58% 45 5.97% 8 14.93% 20 100 134

Police 3.96% 4 42.57% 43 42.57% 6 0.99% 1 44.55% 45 1.98% 2 100 101 World 1.83% 2 13.76% 15 13.76% 41 1.83% 2 2.75% 3 42.20% 46 100 109 Leaders Other 14.61% 13 7.87% 7 7.87% 9 29.21% 26 4.49% 4 33.71% 30 100 89

Anti–corporate globalization protesters: A majority of the combined images (60.5%; N=81) depicted protesters negatively, with 26.9% ( N=36) primary and 33.6% ( N=45) secondary images (see Figure 4). Neutral depictions of protesters ranked second after negative portrayals, with 7.5% ( N=10) primary and 14.9% ( N=20) secondary images neutrally depicting protesters. Positive portrayals of G20 protesters were the least fre - quent, representing 11.2% ( N=15) primary and 6% ( N=8) secondary images.

Police: Overall, an overwhelming majority of all police depictions (87.1%; N=88) were positive, with primary and secondary images constituting 42.6% ( N=43) and 44.6% (N=45), respectively. A low number of images portrayed the police in a neutral fashion, with 5.9% ( N=6) primary and 2% ( N=2) secondary images, respectively. Negative de - pictions of the police were the least frequent, with 4% ( N=4) for primary and 1% ( N=1) for secondary images.

World leaders: News photographs showing world leaders attending G20 meetings were overwhelmingly neutral (79.8%; N=87), of which 37.6% ( N=41) were primary images and 42.2% ( N=46) were secondary. Positive depictions ranked second, with 13.8% (N=15) primary and 2.8% ( N=3) secondary images. Negative images were the least frequent, at 1.8% ( N=2) primary and 1.8% ( N=2) secondary images.

Visual frames and ideology The study selected three influential newspapers from the sample to analyze the rela - tionship between ideological affiliation and visual framing of the G8 and G20 summits. The newspapers’ own editorial policies and leanings were used to classify them in one of the following categories along the Canadian political spectrum: “left of centre” (the Toronto Star ), mainstream political “centre” (the Globe and Mail ), and “right of centre” (the National Post ) (Greenberg, 2000; Jiwani, 2005). A framing analysis of the news - papers’ images covering the G20 protests examined how the visual coverage con - structed one of these overarching “visual frames”: a “lawlessness and violence” frame, a “law and order” frame, or a “civil liberties” frame (see Table 5). Douai “The Police and the Populace” 187

Table 5: Newspapers’ ideological affiliation and visual frames

The Toronto The Globe The National Visual Frame Star and Mail Post % N % N % N “Lawlessness and violence” 27.8 40 21.8 12 36.7 47 “Law and order” 48.6 70 74.6 41 52.3 67 “Civil liberties” 23.6 34 3.6 2 11 14 Total 100 144 100 55 100 128

A first look at the results indicates that the “law and order” frame predominated for the images of all three newspapers, representing 48.6% ( N=70) of the Toronto Star , 74.6% ( N=41) of the Globe and Mail sample, and 52.3% ( N=67) of the National Post sample. while the Globe and Mail G20 images were overwhelmingly depicting the “law and order” frame, Toronto Star ’s images were the least likely to promote this visual frame. A deeper examination of the results reveals that the Toronto Star devoted more space to images that promoted the “civil liberties” frame, with 23.6% ( N= 34), com - pared to the National Post ’s 11% ( N= 14), or the Globe and Mail ’s 3.6% ( N= 2). Moreover, the National Post ’s images were more likely than images from the other two newspa - pers to construct a “lawlessness and violence” frame, with 36.7% ( N= 47), compared to The Toronto Star ’s 27.8% ( N= 40), and the Globe and Mail ’s 21.8% ( N= 12). Visual frames in print versus online mainstream media In addressing the third research question, the study compared mainstream print and online media’s visual framing of the G20 Toronto protests (see Table 6). Despite the differences in the sample, a majority of the images from both print and online media promoted a “law and order” frame. Hence, 59.4% ( N= 249) of the images published in the print media sample clearly privilege the “law and order” frame, and, similarly, 51.9% ( N= 27) of the images published online promoted the same frame.

Table 6: Media type and visual frames

Print Media Online Media Visual Frame % N % N “Lawlessness and violence” 29.8 125 21.2 11 “Law and order” 59.4 249 51.9 27 “Civil liberties” 10.7 45 26.9 14 Total 100 419 100 52

within mainstream media, the visual framing of the protests in print and online sources varies, however, in the degree to which they promoted a “lawlessness and violence” or a “civil liberties” framing of the G20 protests. The “lawlessness and violence” visual frame constituted the second most frequent frame constructed in print media’s images, with 29.8% ( N= 125), while the “civil liberties” frame ranked a 188 Canadian Journal of Communication, Vol 39 (2) distant third at 10.7% ( N= 45). In contrast, the “civil liberties” frame ranked as the second most frequent visual frame to be encountered in online MSM sources, with a 26.9% ( N= 14), while the “lawlessness and violence” frame was the least frequent, with 21.2% ( N= 11). Discussion and conclusions while visual frames work in tandem with textual frames to create an overarching news frame (Mellese & Müller, 2012), this study demonstrates how independent and pow - erful visuals are in framing dissenting voices in the public sphere. The article analyzed a total of N=852 images published in Canadian mainstream media, including eight newspapers and the websites of two leading national news broadcasters. First, the image data indicate an overwhelming “negative” portrayal of the G20 protesters, a very “positive” depiction of the police, and a “neutral” depiction of world leaders and officials attending the summit. Second, the ideological affiliation of newspapers still matters in the visual framing of the anti–corporate globalization movement. However, ideological affiliation of mainstream media fails to translate into a positive coverage of the movement, since the news media are predictably more attracted to shocking images of “lawlessness and violence” compared to those that depict peaceful political dissent or a “civil liberties” frame. Third, a comparison of mainstream media online and print images found little difference between the two, although the “civil liberties” frame was highlighted more in online media compared to print media. The finding that even “left-leaning” media fail to highlight the “civil liberties” frame as prominently as the “lawlessness and violence” frame suggests that the anti–corporate globalization movement and activists should rethink their strategic approach to main - stream media. If anti-globalization protesters’ presumed goal is to “re-appropriate” public space (Juris, 2005) and “hijack” media attention focused on these official summits, they do achieve their goal but at a huge cost to their cause. The larger frame that mainstream media highlight serves little to promote activists’ legitimate concerns about how corpo - rate globalization sidesteps important questions regarding global health and , eq - uitable distribution of economic resources, and democratic governance and human rights issues. The overarching frame, instead, becomes one of “lawless” demonstrators wreaking havoc to the same public space they seek to “liberate.” Hence, visual frames perform ideological functions similar to other media/text frames, in the sense that they both (a) denigrate political dissent while promoting the status quo, and (b) obscure a grassroots critique of social inequalities and economic disparity. The images that capture the media’s imagination are not those of peaceful marches or creative dissent. Juris (2005) explained how the anti–corporate globalization movement’s media coverage re - veals a tension between “traditional” and “non-traditional” forms of protest. On the one hand, traditional forms of protest, such as peaceful street marchers, yield sympathetic, albeit smaller, media coverage. On the other hand, “non-traditional” protests attract greater but less sympathetic media coverage. This tension between forms of protest seems to affect the visual coverage of the G20 protests as journalists, editors, and news organizations are irresistibly attracted to images of chaos and mayhem. These findings go beyond reiterating cumulative research evidence that traditional MSM’s coverage of politics and social movements de-legitimizes dissent (Gitlin, 1980; Douai “The Police and the Populace” 189

McLeod & Hertog, 1992). In fact, the findings cast some doubt on the potential positive impact of new media on the coverage of social movements and political dissent. That images gleaned from mainstream new media sources do not significantly diverge from print media’s images of the G20 protests might be disappointing to those who have hailed the ability of the Internet to challenge the power of mainstream media (e.g., Kahn & Kellner, 2004). This implication has to be qualified, however, because the pres - ent study did not include blogs and social media in its analysis of new media, which should be investigated in future studies. Overall, the study’s main contribution illuminates mainstream media’s visual fram - ing of social protest and anti-globalization movements. The findings demonstrate that ideological leanings, corporate affiliation, and the type of medium have very limited in - fluence on the visual framing of social protest in mainstream media. In this respect, the visual data from different mainstream Canadian media suggest the prevalence of the “lawlessness and violence” frame. The media’s attention is wholly focused on construct - ing a visual narrative of the anti–corporate globalization movement driven by images of private property destruction and anarchy. Alternative explanations of these findings may argue that this attention is driven by newsworthiness considerations and how im - ages of destruction are more sensational than peaceful demonstrations. yet, this inter - pretation does not totally explain how the “civil liberties” frame remains less visible in the image data. The fact that news photographs depicting “civil liberties” frames are difficult to capture should not be used as an a posteriori rationale that exculpates the media from their social responsibility to provide enough visual information on social movements and grassroots protest to counter the media’s excessive focus on pictures of mayhem. Alternative media sources have proven that it is possible to shed light on the “civil liberties” frames and provide more contextual information about the G20 protesters’ social and political platforms (Pickard, 2006; wall, 2003). Despite their importance, these research findings have some limitations. First, the study’s observations about ideological affiliation of the newspaper sample relied on deductive reasoning regarding the editorial policies of the newspapers. This means that a full ideological analysis requires both a close analysis of news reports as well as in-depth interviewing of reporters. Such full ideological analysis, however, falls outside the scope of the present article, because that would require a different set of questions and data gathering resources. Future research should take on this task because it has the potential to enhance visual framing theory. Second, the study’s comparative analy - sis of mainstream print and online media relied on the visual data of two mainstream news websites (CbC.ca and CTV.ca). This was based on the fact that websites of these established, mainstream organizations attract more online visitors than the websites of lesser-known organizations, based on what Hindman (2008) describes as the law of “the winners-take-all” (p. 9). Still, the author acknowledges that a comparative analysis of “alternative” online news sources would shed further light on the potential of online media to challenge the denigration of social protest by hegemonic narratives and ideologies. Moreover, the study recommends that an analysis of the visuals used on television newscasts should be undertaken by future research because it will further enrich our understanding of the visual framing of social movements. 190 Canadian Journal of Communication, Vol 39 (2)

Finally, it is important to reiterate that the prevalence of the “lawlessness and vi - olence” frame in both print and online mainstream media’s photographic coverage of the anti–corporate globalization movement should not be reduced to the media’s pro - fessional structure. It would likewise be simplistic to blame this lopsided visual framing on the ideological affiliation and ownership of the media. In fact, a more balanced ap - proach would recognize that the mainstream media’s professional and ideological lean - ings operate in an information environment in which police services took a more proactive stance in 2010 to de-legitimize and “criminalize” Toronto’s G20 protesters. A public inquiry into Toronto police’s professionalism and questionable conduct re - leased by the Office of the Independent Police review Director (OIPrD) in May 2012 offers some support for this conclusion. based on hundreds of complaints about police tactics and conduct, the OIPrD 2012 investigation acknowledged that “disorganized, poorly trained police had often trampled on the rights of protesters and sometimes resorted to excessive force” (Gee, 2012, n. p.). On the brighter side, and despite their many shortcomings, mainstream media outlets helped document this same heavy- handed policing of citizen protest during the 2010 G20 Summit. References Castells, Manuel. (2007). Communication, power and counter-power in the network society. International Journal of Communication , 1(1), 238–266. De Vreese, Claes H., banducci, Susan A., Semetko, Holli A., & boomgaarden, Hajo G. (2006). The news coverage of the 2004 European Parliamentary election campaign in 25 countries. Politics, 7 (4), 477–504. Dimitrova, Daniela V., Kaid, Lynda L., williams, Andrew P., & Trammell, Kaye D. (2005). war on the web: The immediate news framing of Gulf war II. Harvard International Journal of Press/Politics , 10 (1), 22–44. Entman, robert M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of com - munication , 43 (4), 51–58. Fahmy, Shahira. (2004). Picturing Afghan women: A content analysis of AP wire photographs during the Taliban regime and after the fall of the Taliban regime. Gazette, 66 (2), 91–112. Fahmy, Shahira. (2007). “They took it down”: Exploring determinants of visual reporting in the top - pling of the Saddam Statue in national and international newspapers. Mass Communication & Society , 10 (2), 143–170. Fahmy, Shahira, & Kim, Daekyung. (2008). Picturing the Iraq war: Constructing the image of war in the british and US press. International Communication Gazette , 70 (6), 443–462. Frosh, Paul. (2011). Framing pictures, picturing frames: Visual metaphors in political communica - tions research. Journal of Communication Inquiry , 35 (2), 91–114. Gee, Marcus. (May 16, 2012). report critical of G20 tactics, Chief blair defensive. The Globe and Mail . UrL: http://www.theglobeandmail.com/news/toronto/report-critical-of-g20-tactic-chief- blair-defensive/article4216411 / [accessed June 15, 2012]. Gitlin, Todd. (1980). The whole world is watching: Mass media in the making and unmaking of the New Left . berkeley, CA: University of California Press. Greenberg, Joshua. (2000). Opinion discourse and Canadian newspapers: The case of the Chinese “boat people.” Canadian Journal of Communication , 25 (4), 517–537. Griffin, Michael. (2004). Picturing America’s ‘war on Terrorism’ in and Iraq: Photographic motifs as news frames. Journalism, 5 (4), 381–402. Grunig, Larissa A., Grunig, James E., & Dozier, David M. (2002). Excellent public relations and effective organizations: A study of communication management in three countries . Mahwah, NJ: Lawrence Erlbaum. Douai “The Police and the Populace” 191

Hallahan, Kirk, Holtzhausen, Derina, van ruler, betteke, Verčič, Dejan, & Sriramesh, Krishnamurthy. (2007). Defining strategic communication. International Journal of Strategic Communication , 1(1), 3–35. Hall, Jeanne H., & bettig, ronald V. (2003). “Taking it to the streets”: Mainstream media coverage of anti- global capital protests. Conference papers, International Communication Association, 1–29. Herman, Edward S., & Chomsky, Noam. (1988). Manufacturing consent: The political economy of the mass media . New york, Ny: Pantheon. Hindman, Matthew. (2008). The myth of digital democracy . Princeton, NJ: Princeton University Press. Jay, Martin. (2002). Cultural relativism and the visual turn. Journal of Visual Culture, 1 (3), 267–278. Jha, Sonora. (2008). why they wouldn’t cite from sites: A study of journalists’ perceptions of social move - ment web sites and the impact on their coverage of social protest. Journalism , 9(6), 711–732. Jiwani, yasmin. (2005). The Great white North encounters September 11: race, gender, and nation in Canada’s national daily, The Globe and Mail . Social Justice , 32 (4), 50–68. Juris, Jeffrey S. (2005). The new digital media and activist networking within anti–corporate globalization movements. Annals of the American Academy of Political and Social Science , 597 (1), 189–208. Juris, Jeffrey S. (2008). Performing politics: Image, embodiment, and affective solidarity during anti– corporate globalization protests. Ethnography , 9(1), 61–97. Kahn, richard, & Kellner, Douglas. (2004). New media and Internet activism: From the ‘battle of Seattle’ to blogging. New Media & Society , 6(1), 87–95. Kidd, Dorothy. (2003). Indymedia.org: A new communications commons. In M. McCaughey & M. Ayers (Eds.), Cyberactivism: Online activism in theory and practice (pp. 47–69). New york, Ny: routledge. Lippmann, walter. (1922). Public opinion . New york, Ny: Free Press. Mawby, rob C. (2002). Continuity and change, convergence and divergence: The policy and practice of police-media relations. Criminal Justice, 2 (3), 303–324. McLeod, Douglas M., & Detenber, benjamin. H. (1999). Framing effects of television news coverage of social protest. Journal of Communication, 49 (3), 3–23. Mellese, Mastewal A., & Müller, Marion G. (2012). Mapping text-visual frames of SubSaharan in the news: A comparison of online news reports from and british broadcasting Corporation websites. Communication, Culture & Critique, 5 (2), 191–229. Messaris, Paul, & Abraham, Linus. (2001). The role of images in framing news stories. In Stephen D. reese, Oscar Gandy Jr., & August E. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 215–226). Mahwah, N.J.: Lawrence Erlbaum. Monaghan, Jeffrey, & walby, Kevin. (2012). “They attacked the city”: Security intelligence, the soci - ology of protest policing and the anarchist threat at the 2010 Toronto G20 summit. Current Sociology , 60 (5), 653–671. Parry, Katy. (2010). A visual framing analysis of british press photography during the 2006 - Lebanon conflict. Media, War & Conflict , 3(1), 67–85. Pickard, Victor w. (2006). Assessing the radical democracy of Indymedia: Discursive, technical, and institutional constructions. Critical Studies in Media Communication, 23 (1), 19–38. reese, Stephen. (2001). Prologue –Framing public life: A bridging model for media research. In Stephen reese, Oscar Gandy, and August Grant (Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 7–31). Mahwah, N.J.: Lawrence Erlbaum. Scheufele, Dietram A., & Tewksbury, David. (2007). Framing, agenda setting, and priming: The evo - lution of three media effects models. Journal of Communication, 57 (1), 9–20. Schwalbe, Carol b. (2006). remembering our shared past: Visually framing the Iraq war on U.S. news websites. Journal of Computer-Mediated Communication, 12 (1), 264–289. UrL: http://jcmc.indiana.edu/vol12/issue1/schwalbe.html [accessed January, 20, 2010] Schwalbe, Carol b., Silcock, b. william, & Keith, Susan. (2008). Visual framing of the early weeks of the US-led invasion of Iraq: Applying the master war narrative to electronic and print im - ages. Journal of Broadcasting & Electronic Media , 52 (3), 448–465. Sontag, Susan. (2004, May 23). regarding the torture of others. The New York Times Sunday Magazine , pp. 25–29. 192 Canadian Journal of Communication, Vol 39 (2)

Thompson, John b. (2000). Political scandal: Power and visibility in the media age . Cambridge, UK: Polity Press. Toronto Police Service (TPS). (2011, June). G20 Summit Toronto, Ontario, June 2010: Toronto Police Service After-Action review, June 2011. UrL: http://www.torontopolice.on.ca/publica - tions/files/reports/g20_after_action_review.pdf [accessed July 2, 2012] wall, Melissa. (2003). Press conferences or puppets: NGOs’ vs. street groups’ communication in the battle of Seattle. Javnost—The Public , 10 (1), 33–48.